Machine learning-based decomposed fuzzy set model for analyzing key performance indicators in the waste-to-energy supply chain

Abstract

Waste management through circular economy implementation is crucial for achieving sustainability and enhancing the performance of the waste-to-energy supply chain (WtESC). Therefore, developing key performance indicators (KPIs) and understanding their significance is essential for assessing WtESC performance. However, there is a lack of studies focused on developing and evaluating KPIs for WtESC. To address this gap, this study offers a novel machine learning (ML)-based decomposed fuzzy set (DFS)-analytical hierarchy process (AHP) model to assess the KPIs that can be used to evaluate the WtESC performance. Since decision-making based on experts’ judgment often faces uncertainty and experts’ experience significantly impacts the final decision, the advanced ML-based DFS-AHP model can effectively handle these challenges and enhance the model’s reliability. In the proposed framework, decision makers’ weights are computed using the ML approach based on expert information, which is integrated into the DFS-AHP model. The results indicate that the most important KPI for WtESC is ‘CO2 emissions intensity’, which received a de-fuzzified composite weight of 0.1360. This KPI should be considered with a higher priority to ensure sustainability and improve WtESC performance. Consequently, the decision-makers should consider these findings when developing the performance index for WtESC, which may further assist in taking the necessary actions to improve WtESC’s performance.